Project 1 - Terrorists

SSIP 2005

Aim

• Terrorists need to be identified when passing a security screen
  

• Aim is positive identification of a few faces
  

• Problem is that terrorists try to disguise themselves

• Finding the terrorists
  

• Identifying faces even if they are disguised
  

Method

Step 1

• Getting the images/photos

• Detecting eyes by Isis algorithm

• Transforming the eyes onto a horizonal line

• Normalizing the images (to see just the face), and placing the eyes into correct position so they are unified

Step 2

• Giving the pictures to a learning algorithm. We use the Eigenfaces

Step 3

• From these normalized images the Eigenface algorithm detects and classifyies the input picture with the help of it's databas

  ---

Implementation

• Programmed in Matlab
  
  
• Preprocessing
  
  Finding eyes
  
  Rotating
  
  Scaling
  

  Images normalized to 128*128

   

• Processing
Learning algorithm – Eigenfaces

Creating a database

Classifying – Eigenfaces algorithm